Clouds play a critical role in climate change and the energy budget in the Arctic. The parameterization of cloud microphysics is crucial for simulating surface radiation and atmospheric vertical structure in the Arctic region. This study evaluates the performance of eight widely used cloud microphysics schemes in the Svalbard region during winter and summer, utilizing field observation data and the polar-optimized Weather Research and Forecasting model (Polar-WRF). The results demonstrated that the choice of cloud microphysics scheme significantly influences surface radiation and simulations of the lower atmosphere. During winter, the Thompson, Morrison, and NSSL schemes effectively simulate surface radiation, whereas the Lin, Goddard, and WSM6 schemes excel in capturing the vertical structure of the atmosphere. In summer, the WSM6 and Thompson schemes show superior performance in simulating surface radiation, whereas the Thompson, Morrison, and CAM5.1 schemes better represent atmospheric profiles. The accuracy of cloud simulations is crucial for surface radiation, as all schemes tend to underestimate downward longwave radiation, primarily due to an overestimation of low cloud cover. In addition, challenges in simulating mixed-phase clouds, which are prevalent in the Svalbard region, were identified, with most schemes overestimating the ice water content and underestimating the liquid water content in winter. In summer, models tend to underestimate the cloud height and liquid water content of low-level clouds. These findings highlight the importance of accurately simulating low-level mixed-phase clouds to improve the understanding of surface–atmosphere interactions in the Arctic.